Re: result-document, QNames, AVT's, format-attribute a

> Doesn't that struck you as odd?

Even Michael's allowed the occasional bug.. (See his reply)

> They should've asked me in the board and I certainly would've convinced 
> them otherwise ;-)

The Working drafts have been open for public comment for half a decade
or so:-)

I couldn't actually find an issue raised against unparsed text but there
are some against codepoint-to-string (which is the same issue, teh data
model doesn't allow the characters, so the functions can not)

most recently
http://www.w3.org/Bugs/Public/show_bug.cgi?id=3838
I'm sure that I made a comment much earlier, but I don't see it now.

David